Zhang Ye1,2, Dina M. Silva2, Daniela Traini2,3, Paul Young2,4, Shaokoon Cheng1, Hui Xin Ong2,3
1 School of Mechanical Engineering, Faculty of Engineering, Macquarie University, Sydney, NSW, Australia
2 Woolcock Institute of Medical Research, Sydney, Australia.
3 Department of Biomedical Sciences, Faculty of Medicine, Health and Human Sciences, Macquarie University, Sydney, NSW, Australia
4 Department of Marketing, Macquarie Business School, Macquarie University, Sydney, NSW, Australia
Summary
Pseudomonas aeruginosa (P. aeruginosa) biofilm colonizing and growing in the human respiratory tract is a known cause of reduced antimicrobial response in several chronic respiratory diseases. Although numerous in vitro models have been developed to provide important insights about biofilm structure and promising treatments for biofilm-related infections, most models are still based on mono-interface culture. The handful of air-liquid interface (ALI) models developed, are incapable of manipulating the airflow dynamics, an essential feature to mimic different respiratory regions and disease conditions. The failure to reproduce the host environment may lead to misleading results. In this study we developed a dual-chamber microfluidic device and set up a dynamic platform capable of establishing an ALI model to closely mimick the lung environment. Using this platform, 48 h old P. aeruginosa biofilms were cultured, and their development studied as a function of nutrient supply conditions, in addition to aerodynamic shear forces. The biofilm samples were investigated in a cross-sectional study to compare their viable cell number, morphology, antibiotic susceptibility to a model antiobiotic, ciprofloxacin hydrochloride (CIP), and biofilm matrix permeability.
Key Message
Our study shows that biofilms developed in nutrient-rich conditions are thicker, less permeable, and more resistant to antibiotics compared to those grown in nutrient-depleted conditions. Under nutrient-rich conditions, mechanical shear forces induced from airflow dynamics produced thinner and less permeable biofilms, but no apparent alternation in antibiotic susceptibility were observed. Moreover, it was found that the minimum biofilm eradication concentration (MBEC) of CIP using our device was significantly higher than the conventional microtiter plate method. These findings indicate that using less physiologically relevant biofilm in vitro models could lead to an overestimation of drug efficacy, potentially leading to clinical failure.